Go to PegaPoll
PEGAPOLL · NEWS

Google’s Gemini 4 Neon is faster, cheaper and smarter than Carbon

The Decoder (AI) · 2026-11-19
🤖 AI-generated content — The title and summary were produced automatically by artificial intelligence, without human editorial review.

Google DeepMind has introduced Gemini 4 Neon, its fourth AI model in seven weeks. Positioned between Gemini 4 Carbon and Hydrogen, Neon combines a one-million-token context window for text, images, audio and video with substantially lower pricing and faster processing than Carbon. At its highest “xhigh” reasoning level, it scores 61 on the Intelligence Index, one point above Carbon’s 60. Google’s comparison omits GPT-6.2 Astra and Claude Fable 5.2, which rank ahead of Neon in the article’s wider comparison.

Through Google AI Studio and Gemini for Business, Neon costs $3 per million input tokens and $15 per million output tokens, or $0.15 for cached input. Those rates are 25% below Carbon’s regular $4 and $20 prices, though Carbon’s promotional pricing remains lower through the end of November. Artificial Analysis estimates Neon’s xhigh cost per task at $6.44, compared with $7.76 for Carbon, while its processing time falls from about 36 minutes to 21 minutes. Neon generates roughly 110 tokens per second, nearly twice Carbon’s 57. It also performs strongly in coding, visual document analysis, presentation quality and security evaluations, including a reported 90.5% on SEC-bench Pro for its cyber variant.

Continue in the app — vote & join in ➔
Source: The Decoder (AI) · via ahirlevel.hu

Related news